Moore’s law and the physical limits of computing

In 1959, Douglas Engelbart studied the projected downscaling of integrated circuit (IC) size, publishing his results in the article "Microelectronics, and the Art of Similitude". Engelbart presented his findings at the 1960 International Solid-State Circuits Conference, where Moore was present in the audience.

In 1965, Gordon Moore, who at the time was working as the director of research and development at Fairchild Semiconductor, was asked to contribute to the thirty-fifth-anniversary issue of Electronics magazine with a prediction on the future of the semiconductor components industry over the next ten years. His response was a brief article entitled "Cramming more components onto integrated circuits". Within his editorial, he speculated that by 1975 it would be possible to contain as many as 65000 components on a single quarter-square-inch (~ 1.6 cm) semiconductor.

The complexity for minimum component costs has increased at a rate of roughly a factor of two per year. Certainly over the short term this rate can be expected to continue, if not to increase. Over the longer term, the rate of increase is a bit more uncertain, although there is no reason to believe it will not remain nearly constant for at least 10 years.

Moore posited a log–linear relationship between device complexity (higher circuit density at reduced cost) and time. In a 2015 interview, Moore noted of the 1965 article: "... I just did a wild extrapolation saying it's going to continue to double every year for the next 10 years." One historian of the law cites Stigler's law of eponymy, to introduce the fact that the regular doubling of components was known to many working in the field.

In 1974, Robert H. Dennard at IBM recognized the rapid MOSFET scaling technology and formulated what became known as Dennard scaling, which describes that as MOS transistors get smaller, their power density stays constant such that the power use remains in proportion with area. Evidence from the semiconductor industry shows that this inverse relationship between power density and areal density broke down in the mid-2000s.

At the 1975 IEEE International Electron Devices Meeting, Moore revised his forecast rate, predicting semiconductor complexity would continue to double annually until about 1980, after which it would decrease to a rate of doubling approximately every two years. …

Numerous innovations by scientists and engineers have sustained Moore's law since the beginning of the IC era. Some of the key innovations are listed below, as examples of breakthroughs that have advanced integrated circuit and semiconductor device fabrication technology, allowing transistor counts to grow by more than seven orders of magnitude in less than five decades.

Computer industry technology road maps predicted in 2001 that Moore's law would continue for several generations of semiconductor chips.

One of the key technical challenges of engineering future nanoscale transistors is the design of gates. As device dimensions shrink, controlling the current flow in the thin channel becomes more difficult. Modern nanoscale transistors typically take the form of multi-gate MOSFETs, with the FinFET being the most common nanoscale transistor. The FinFET has gate dielectric on three sides of the channel. In comparison, the gate-all-around MOSFET (GAAFET) structure has even better gate control.

Microprocessor architects report that semiconductor advancement has slowed industry-wide since around 2010, below the pace predicted by Moore's law. Brian Krzanich, the former CEO of Intel, announced in 2015, "Our cadence today is closer to two and a half years than two." Intel stated in 2015 that improvements in MOSFET devices have slowed, starting at the 22 nm feature width around 2012, and continuing at 14 nm. Pat Gelsinger, former Intel CEO, stated at the end of 2023 that "we're no longer in the golden era of Moore's Law, it's much, much harder now, so we're probably doubling effectively closer to every three years now, so we've definitely seen a slowing."

The physical limits to transistor scaling have been reached due to source-to-drain leakage, limited gate metals and limited options for channel material. Other approaches are being investigated, which do not rely on physical scaling. These include the spin state of electron spintronics, tunnel junctions, and advanced confinement of channel materials via nano-wire geometry. Spin-based logic and memory options are being developed actively in labs.

The vast majority of current transistors on integrated circuits are composed principally of doped silicon and its alloys. …

In April 2005, Gordon Moore stated in an interview that the projection cannot be sustained indefinitely: "It can't continue forever. The nature of exponentials is that you push them out and eventually disaster happens." He also noted that transistors eventually would reach the limits of miniaturization at atomic levels:

In terms of size [of transistors] you can see that we're approaching the size of atoms which is a fundamental barrier, but it'll be two or three generations before we get that far—but that's as far out as we've ever been able to see. We have another 10 to 20 years before we reach a fundamental limit. By then they'll be able to make bigger chips and have transistor budgets in the billions.

In 2016 the International Technology Roadmap for Semiconductors, after using Moore's Law to drive the industry since 1998, produced its final roadmap. It no longer centered its research and development plan on Moore's law. Instead, it outlined what might be called the More than Moore strategy in which the needs of applications drive chip development, rather than a focus on semiconductor scaling. Application drivers range from smartphones to AI to data centers.

IEEE began a road-mapping initiative in 2016, Rebooting Computing, named the International Roadmap for Devices and Systems (IRDS).

Some forecasters, including Gordon Moore, predicted between 2012–2016 that Moore's law would end by around 2025. Although Moore's Law will reach a physical limit, some forecasters in 2019 and 2020 were optimistic about the continuation of technological progress in a variety of other areas, including new chip architectures, quantum computing, and AI and machine learning. Nvidia CEO Jensen Huang declared Moore's law dead in 2022; several days later, Intel CEO Pat Gelsinger countered with the opposite claim.

Digital electronics have contributed to world economic growth in the late twentieth and early twenty-first centuries. The primary driving force of economic growth is the growth of productivity, which Moore's law factors into. Moore (1995) expected that "the rate of technological progress is going to be controlled from financial realities". The reverse could and did occur around the late-1990s, however, with economists reporting that "Productivity growth is the key economic indicator of innovation." Moore's law describes a driving force of technological and social change, productivity, and economic growth.

An acceleration in the rate of semiconductor progress contributed to a surge in U.S. productivity growth, which reached 3.4% per year in 1997–2004, outpacing the 1.6% per year during both 1972–1996 and 2005–2013. As economist Richard G. Anderson notes, "Numerous studies have traced the cause of the productivity acceleration to technological innovations in the production of semiconductors that sharply reduced the prices of such components and of the products that contain them (as well as expanding the capabilities of such products)."

The primary negative implication of Moore's law is that it is associated with rapid obsolescence and accordingly high maintenance costs. As technologies continue to rapidly improve, they render predecessor technologies obsolete. In situations in which security and survivability of hardware or data are paramount, or in which resources are limited, rapid obsolescence often poses obstacles to smooth or continued operations.

Several measures of digital technology are improving at exponential rates related to Moore's law, including the size, cost, density, and speed of components. Moore wrote only about the density of components, "a component being a transistor, resistor, diode or capacitor", at minimum cost.

Transistors per integrated circuit – The most popular formulation is of the doubling of the number of transistors on ICs every two years. At the end of the 1970s, Moore's law became known as the limit for the number of transistors on the most complex chips. The graph at the top of this article shows this trend holds true today. As of 2025, the commercially available processor possessing one of the highest numbers of transistors is a GB202 graphics processor with more than 92.2 billion transistors.

Density at minimum cost per transistor – This is the formulation given in Moore's 1965 paper. It is not just about the density of transistors that can be achieved, but about the density of transistors at which the cost per transistor is the lowest.

As more transistors are put on a chip, the cost to make each transistor decreases, but the chance that the chip will not work due to a defect increases. In 1965, Moore examined the density of transistors at which cost is minimized, and observed that, as transistors were made smaller through advances in photolithography, this number would increase at "a rate of roughly a factor of two per year".

Dennard scaling – This posits that power usage would decrease in proportion to area (both voltage and current being proportional to length) of transistors. Combined with Moore's law, performance per watt would grow at roughly the same rate as transistor density, doubling every 1–2 years. According to Dennard scaling transistor dimensions would be scaled by 30% (0.7×) every technology generation, thus reducing their area by 50%. This would reduce the delay by 30% (0.7×) and therefore increase operating frequency by about 40% (1.4×). Finally, to keep electric field constant, voltage would be reduced by 30%, reducing energy by 65% and power (at 1.4× frequency) by 50%. Therefore, in every technology generation transistor density would double, circuit becomes 40% faster, while power consumption (with twice the number of transistors) stays the same. …